I’ve been thinking seriously about description for some time now, and in distinguishing it from interpretation and/or explanation. If we then add evaluation to that we have the four methodological roles that Sharon Marcus investigates in “Erich Auerbach’s Mimesis and the Value of Scale” (Modern Language Quarterly 77.3, 2016, 297-319).
As the result of a Twitter conversation initiated yesterday by Ted Underwood, and to which Katherine Bode contributed in an important way (see Appendix 1), I now realize that that’s just what those words designate, roles, roles in a discourse. While that may seem obvious, it’s an important realization (for me). For this line of investigation began with the desire to think of description as a relatively transparent intellectual act, in distinction from interpretation or explanation, which are difficult and often (deeply) problematic.
The purpose of this post is to begin thinking about what those roles are. In the next section I do that briefly. Then I have a section about description where I talk about the point of thinking of these activities as roles in a discourse. Then I quote passages from Marcus’s article and add a bit of light commentary. I conclude with one appendix pointing to the Twitter discussion I’ve already mentioned and another appendix with links to some of my work on description.
Four discourse roles
Description and evaluation seem to me to be the easiest to specify.
Description sets the terms in which a phenomenon is introduced into discourse and presented for investigation through any of the other three roles. As I am particularly interested in literary phenomena I am particularly interested in the description of texts. In particular, I am interested in describing the formal features of texts.
It is through evaluation the phenomena are related to vital human interests. This is the realm of ethics and aesthetics, matters that have been de-emphasized in literary theory, though they pervade the practice of criticism.
Specifying the role or roles of interpretation and explanation seems a bit trickier. For now I think it’s sufficient to think of it as involving something like comprehension or understanding. It is through interpretation or explanation that we link described phenomena to something else. Interpretation is about meaning while explanation is about mechanism. Interpretation is ultimately subjective while explanation aspires to objectivity.
That, I know, is vague, and even problematic. But that’s OK. I’m just trying to get things started, not get them all sorted out.
[Interpretation links description to meaning and evaluation in ethical criticism. Explanation links description of texts to psychological, neural, and social mechanisms.]
I can do better on both interpretation and explanation, and evaluation for that matter,
but I'll save that for later posts rather than revising this one. For an idea of what I have in mind, see the note, The rhetoric of interpretation and explanation in literary criticism, in this post.
But why roles? The case of description
As I’ve indicated, this line of thought began with a desire to think of description as a transparent process. Why? Because I’m interested in objectivity. I wanted to extend the range of objective truths, if you will, that we can state about literary texts. I figured that, if description were a transparent process, that would be the way to do it. But Stanley Fish talked me out of it. So distinguishing description from something else, like interpretation or explanation, became my fallback position.
There’s no need to go through that (you can find some of the reasoning in working papers listed in Appendix 2). And so I continued on, now thinking about biology. Biology’s a discipline based on hundreds of thousands of descriptions of life forms and their lifeways. That’s what Charles Darwin had available to him when he began thinking about evolution. Those descriptions were mostly based on naked eye observations that became organized through reference collections, verbal descriptions, and drawings.
Then, in the 20th century, molecular biology emerged; it deals with objects too small for naked eye observation. They can’t even be seen in a microscope. Watson and Crick got a Nobel Prize for discovering that the DNA molecule has the form of a double helix. That’s a descriptive characterization (their 1953 paper in Nature also had a simple drawing). But the process through which they arrived at that description was hardly a direct and transparent one. It involved a sophisticated observational technique (X-Ray crystallography) and mathematical modeling and inference.
So, description, mere description, can be a complex and sophisticated process, one fraught with opportunities for error and confusion.
But what does this have to do with literary criticism? you may ask. Consider computational criticism (aka ‘distant reading’). It seemed obvious to me, and to others, that the charts and graphs that are typically used to present the computational results are fundamentally descriptive in character. They don’t give us the meaning of a corpus–a question to which Alan Liu has given much attention–nor do they explain anything. And they certainly don’t take the aesthetic or ethical temperature of a body of literary work. They just describe features of those texts considered as a single object, a corpus.
Those descriptive objects are not easy to understand. And the processes that produce them are complex and problematic. They can easily go astray.
By saying that description is a role, we can treat the process by which those charts and graphs are produced has something happening within a ‘black-box’. We can then take those descriptions at face value for purposes of further reflection and investigation. What goes on inside that black box, however, will involve its own processes of description, interpretation, explanation and, yes, evaluation.
Thus when one says that ‘distant reading’ produces descriptive charts and graphs, one isn’t characterizing them as objective ‘facts’. They may or may not be; but that’s a separate discussion. All one is doing is asserting that they play a certain role in a larger conversation.
How Marcus characterizes these four roles
Here are some passages from her article, “Erich Auerbach’s Mimesis and the Value of Scale”, along with some light commentary.
Description states what things are and how they work; according to OED2 (www.oed.com, accessed April 26, 2016), the verb describe means “to portray in words or by visual representation” and “to give an account of or statement about in speech or writing.” While devalued and controversial among literature scholars, description remains a common technique in music and art history, where it is seen as requiring training and erudition. In literary studies, description takes center stage in philology, narratology, poetics, stylistics, bibliography, and book history, as well as in some aspects of the digital humanities, such as ontologies, text mining, and text encoding. Many critics negatively associate description with tautology (see, e.g., Rooney 2010). Description does rely on sanctioned forms of tautology, such as quotation, but description also involves acts of categorization and classification that usually generate a vocabulary more abstract than that of the objects it analyzes. Thus many descriptive statements in literary criticism correlate specific textual features to terms drawn from grammar, rhetoric, genre criticism, narratology, and history.
I note that Marcus treats description as purely/primarily verbal, which is common in recent critical discussions of description. But visual objects can also be descriptive in character. That is certainly that case in biology and now in distant reading.
But that is also true in linguistics, for example, where tree diagrams are common. And then we have the various tables, diagrams, and quasi-mathematical formulas Lévi-Strauss used in his analysis of ethnographic materials, including, of course, his analysis of myths. These all have a descriptive character – a matter I discuss in my working papers (Appendix 2).
Finally, in Reassembling the Social (2005) Bruno Latour asserts (136): “No scholar should find humiliating the task of description. This is, on the contrary, the highest and rarest achievement.”
Interpretation states not what things are or how they work but what they mean. The boundaries between description and interpretation are contested and blurry, and I do not propose to settle them here. In everyday speech, interpretation is often synonymous with opinions based on beliefs that cannot be proven, on flimsy reasoning or evidence, or on views that are self-servingly biased (“That’s just your interpretation”). In literary criticism, there are two common ways to distinguish interpretation from description. One is to align description with statements that claim indisputability and interpretation with statements that avow their partiality. Since most interpretations (even avowals of interestedness) carry truth claims and most descriptions are incomplete or situated, I do not consider this a useful distinction. Instead, I would propose that the two terms exist on a spectrum. We move closest to the interpretive end of the spectrum when we argue that the text means something very different from what it says or when we assert that a text’s meaning and import lie in what it does not say, in blind spots and exclusions that only the interpreter can point out.
I note that the distinction between description and interpretation is problematic. That’s an issue. I have no reason to think that characterizing description and interpretation as roles settles the issue, but it may be a useful way of thinking about what’s going on here.
It gives the matter a different valence, if you will. It is no longer a matter of figuring out what a statement is – description or interpretation? – but of asking how it functions and whether or not it performs that role well. A given statement may well perform two (or even more) roles.
You may also wish to consider a point that Michael Bérubé makes in “There is Nothing Inside the Text, or, Why No One’s Heard of Wolfgang Iser”, (Postmodern Sophistry: Stanley Fish and the Critical Enterprise, edited by Gary A. Olson and Lynn Worsham, SUNY Press 2004, pp. 11-26).
... It would have been possible, in other words, to contest Fish’s reading of Iser [...] by acknowledging that all forms of reading are interpretive but that some involve the kind of low-level, relatively uncontestable cognitive acts we engage in whenever we interpret the letter “e” as the letter “e,” and some involve the kind of high level, exceptionally specific and complex textual manipulations, transformations and reconfigurations involved whenever someone publishes something like S/Z – or Surprised by Sin. (And, of course, that there are any number of “interpretations” that fall between these extremes, and that the status of each of them is – what else? – both open to and dependent on interpretation.)
Explanation designates the operation by which literary critics assign causality, though explanation can also signify description and interpretation, as when we “explain” a poem. Literary critics tend to downplay causality — “why?” is not our favorite question — and usually refer the sources of a text’s meaning or form to disciplines other than literary criticism, such as history, biography, economics, philosophy, or neuroscience. Thus scholars often relate specific features of literary works to general phenomena such as modernity, capitalism, imperialism, patriarchy, or the structure of our brains. But because explanation is an undervalued operation in literary criticism, one seen to depend on the kind of literalism that leads many critics to reject description as impossible, the exact nature of the link between general phenomena and specific works often remains nebulous. Literary critics are more likely to posit the relationship between the realist novel and capitalism as one of homology, analogy, or shared commitments (to, say, individualism) than they are to trace a clear line from one as cause to the other as effect.
One might think of interpretation and explanation as alternative ways of offering causal accounts. Classically (if I may) in literary criticism, the cause of a text is said to be the author. The classical critic seeks the author’s intention as the source of a text’s meaning. Why is the text what it is? How did it come into being? The author did it.
Post-classically, the author gets bracketed out in favor of social, semiotic, and psychological forces operating through the author. It is those forces that bring the text into existence and are the source of its meaning. The post-classical critic then smuggles evaluation in by way of critique, thereby completing the circuit and linking criticism to those existential concerns – what is the good? how do I live? – that motivate literature itself.
Evaluation involves assessment, appraisal, and judgment of importance, merit, quality, and social and political effects. Evaluation can tell us why a text is good or bad, succeeds or fails, is worth reading or not. Until the 1960s academic literary critics often engaged in explicit aesthetic evaluation, but since then assessment of texts’ readability or worth has migrated primarily to book reviews or to the more implicit operations of canon formation and curriculum production. Evaluation remains a strong force in literary scholarship, however, in the form of critique: disapproval of or dissent from what a text says or means, usually on ethical or political grounds. Critique can easily be knit into description, interpretation, and explanation: one can disapprove of what a text does or does not describe, dissent from what it means, protest its underlying causes, or do all three at once.
I note that the educated public wants literary critics to evaluate texts. I also think that we should explicitly and deliberately theorize it and bring it back as a critical activity. Yes, as Marcus notes, it’s there in critique, but we don’t really want to own up to what we are doing. We take the values embodied in critique at face value and so place this or that critical stance itself being above criticism.
Appendix 1: A twitter conversation
Good reflection on explanatory problems in distant reading. Short take-away? "These are difficult problems." https://t.co/9C6mQFgOsa— Ted Underwood (@Ted_Underwood) August 8, 2017
Katherine Bode joined in here:
and @bbenzon where is the line between description and explanation? At the stage of describing data I've already interpreted it many times— Kath Bode (@KatherineBode) August 8, 2017
That phase of the conversation ended here:
U have no idea how much blood I shed learning that "description" is a role, not more or less a synonym for "fact".— Bill Benzon (@bbenzon) August 9, 2017
Others have picked up a thread and continued the conversation today (August 9, 2017).
Appendix 2: Some of my work on description
I have written quite a bit about description in the past few years. At the moment (8.9.2017) I have 129 blog posts, including this one, at New Savanna. I have collected some of them into three working papers. You can find them at Academia.edu at this link. Individually they are:
This is a series of notes in which I argue that better descriptive methods are a necessary precondition for more sophisticated and objective literary criticism. Description, though it does not give unmediated access to texts, requires methods for objectifying texts, methods which must be discovered in the doing. By way of comparison I discuss the role of description in biology and I discuss the use of images and diagrams as descriptive devices. Lévi-Strauss on myth and Franco Moretti on distant reading, though quite different, are up to the same thing: objectification.
Description 2: The Primacy of the Text, October 2013.
These notes consist of five posts discussing the description of literary texts (including comparative analysis of individual texts and topic modeling) and films plus four appendices containing tables used in describing two manga texts (Lost World, Metropolis) and two films (Sita Sings the Blues, Ghost in the Shell 2: Innocence). The posts make the point that the point of description is to let the texts speak for themselves. Further, it is through descriptions that the texts enter intellectual discourse.
Description 3: The Primacy of Visualization, October 2015.
Describing literary texts requires a mode of thought distinct from the discursive interpretation of them. It is a mode of thought in which various visual devices are central. These devices include: tables, trees and mental spaces, directed graphs and “sketchpads”. Visualization facilitates the objectification of literary form and objectification is necessary for objectivity. With objectivity comes the possibility of cumulative knowledge.